FitBenchmarking review
A focused open-source toolkit for benchmarking nonlinear fitting algorithms.
Reviewed by iTechGuides Editors · Editorial team · Updated Oct 2026
FitBenchmarking is an open-source Python package for scientists and scientific or numerical software developers comparing nonlinear least-squares fitting software and minimizers. It lets users benchmark algorithm performance across fitting problems and hardware, making it a fit for researchers evaluating optimization methods rather than teams looking for a general statistical analysis suite. The package runs on Windows, macOS and Linux and can be self-hosted.
Its strength is configurable benchmarking. Users can select fitting software and algorithm types, choose nonlinear least-squares cost functions—including weighted and Poisson options—and select Jacobian and Hessian methods. Repeated runs and maximum runtime can also be configured. Supplied benchmark problems provide a starting point, while custom problem sets are supported through documented formats including CUTEst, Native, Mantid, NIST, Horace, HOGBEN and BAL. Results can be browsed in tables and interactive plots through a Dash app. Integrations include Bumps, DFO-LS, Gradient-Free-Optimizers, Minuit, SciPy, Mantid, MATLAB and SASmodels.
FitBenchmarking is free and open source, with no paid plan tiers; it is a package rather than a hosted commercial service. Its specialization is also its boundary: it focuses on comparing minimizers and fitting frameworks, and does not provide statistical tests or regression models. That makes it less suitable for analysts seeking broad statistical workflows or a conventional end-to-end analysis environment. The project is developed through collaboration between STFC's Scientific Computing Department, ISIS Neutron and Muon Facility, and Diamond Light Source, with community support. Researchers who need repeatable comparisons of nonlinear fitting approaches should consider it; those needing broader statistical analysis should look elsewhere.
FitBenchmarking pros and cons
- Where it wins
- Compare minimizers and fitting frameworks on common benchmark problems
- Configure cost functions, derivative methods, repeated runs and runtime limits
- Inspect tabular results and interactive plots in a Dash app
- Where it doesn't
- Focused on optimization benchmarking, not end-to-end statistical analysis
- Does not provide statistical tests or regression models
- Support is community-based
FitBenchmarking fact sheet, pricing and score →
Advertiser disclosure: iTechGuides is reader-supported. Vendors can pay for top positions in our rankings and for a place on other products' pages, and we may earn a commission when you click some links. Bought places stand at the number bought; this page and our disclosure say so. Paying vendors may get a fuller, more detailed review from our editors, and that review can change their score. How we rank.
Last updated · How we research and update